Where Have All The Cycles Gone?

Mai writes "Computers are getting faster all the time, or so they tell us. But, in fact, the user experience of performance hasn't improved much over the past 15 years. This article takes a look at where all the precious processor time and memory are going."

Comments

Code Bloat

[RatBastard]

Simple. Code bloat.

Re:Code Bloat

[pilkul]

Or in other words, developers spending less time on petty optimization to work on features and bugfixes.

Re:Code Bloat

[Anonymous Coward]

You want cross platform, you want platform support, so get used to VMs and WRAPPER code.

You want language interop, you have to live with COM, DCOM, CORBA, XML Web services, SOAP, UNO. If not, then go back to your Amstrad or Sinclair.

FFS you cant have your cake and eat it.

Re:Code Bloat

[Rei]

More than just code bloat: disk access speeds combined with the increase in size of everything (for "richer content", "more features", etc).

With a lot more files, a lot larger files, etc, your performance at file access (and disk caching) will decrease unless you can increase the throughput and decrease the seek time/latency of data access from the disk.

Back in 1998, most PCs were shipping with 5400 RPM hard drives; I bought a used 10k rpm drive for 200$. Nowadays? Most PCs are still shipping with 5,400 RPM drives, and 10k RPM is still good performance (although not the best out there - although, mine back then wasn't top of the line either).

Disk sizes are scaling up wonderfully. Disk access speeds are not. The same holds true with RAM.

Re:Code Bloat

[pilkul]

The bottom line is that optimization is rarely the best thing a developer can be doing with his time. There's a lot more you can do with "a day" than tweak a poorly thought out algorithm. By all means optimize if your program is as slow as molasses and your profiling shows that a particular section of your code is at fault. Otherwise, good programming practice is to think about other things.

Just look at the size of a word document today

[marika]

I used to be able to fit the system, the finder, mac write and 3 X 8K docs on a floppy. Now I could barely fit a word document.

Re:Just look at the size of a word document today

[bongoras]

But you can still fit an entire book on a floppy if you use LaTeX. The morale of the story: Don't want a slow, bloated system? Then tough it out and don't use one. But don't complain when you have to type:

\begin{enumerate}
\item Open a terminal window by right clicking on your desktop and selecting ``Open Terminal''
\item In that window, become root by typing {\tt su}
\item Now put a blank CD in your drive and burn the iso image to it by typing \\
{\tt cdrecord -dev=0,0,0 cdimageyouwanttoburn.iso}
\end{enumerate}

instead of clicking the bullet button or asking a paperclip to make a list. It's all a matter of what you want. There are plenty of lean, mean systems out there. Don't bitch about UI slowness unless you are willing to use a plain-text console with "screen" "mutt" and "elinks" as your main applications.

Re:Just look at the size of a word document today

[squiggleslash]

Wow, talk about extremes.

I'm pretty sure there's a half-way house somewhere. I seem to remember the early nineties were full of fully functional does-everything-I-want applications that had nice GUIs and used a megabyte or two of memory when they ran.

I wouldn't categorize MacWrite or early Mac OS as that, but, I don't know, Lotus AmiPro and Microsoft Word for Windows 2 were pretty good, to give two examples.

Re:Just look at the size of a word document today

[PingPongBoy]

Don't forget page formatting, font formatting, paragraph formatting, printer settings, etc. etc.

Go back to Data Structures 101. A lot of data structures have a head node that don't contain useful data but serve as a place to find the structure. With all the objects that can be put into a Word doc, you can expect a lot of head nodes.

Most of the time you don't save a Word doc with just one character. You are working on a massive doc that you save frequently but you don't want to wait for the whole thing to be written every time you save. There has to be some overhead for faster saves.

Re:Just look at the size of a word document today

[batemanm]

I think Word has actually gotten better in terms of file size, it is still a complete pain in the arse and the bain of my life. When I did my undergraduate degree the report was around 10 meg, that was on Word 6.0. It was around 100 pages and have very few pictures. Now my PhD thesis is around 300 pages and is full of pictures and is currently about 6.5 meg (Word 2003). I have saved my PhD thesis out to Word 6 format before (to fix problems that word was having) and the resulting file was over 63 meg.

Where are all the precious cycles going?

[R2.0]

spyware, trojans, p2p apps.

Nobody give a fig about optimizing

[winkydink]

'memory is cheap'
'disks are fast'
'processors are fast'

nobody cares about optimizing code anymore.

Re:Nobody give a fig about optimizing

[cortana]

That's just what the assembley programmers said when researchers were moving to C. ;)

Re:Nobody give a fig about optimizing

[dattaway]

Back in the old days, popular applications were stand alone, written in assembly, and made to fit within a single code segment. Since resources were small, much care was taken to get the most out of so little. Software back then were simple like motorcycles; they had the basics bolted to a simple frame and off it went. Today we have software written with stock libraries, made to work with all kinds of resources and standards, and required to work with large filesystems and memory maps. Applications back then fly today, but seem like a small insect when it comes to functionality.

Re:Nobody give a fig about optimizing

[tomstdenis]

I'd hope not because for most problems efficient *algorithms* are more important than efficient implementation.

Tom

Re:Nobody give a fig about optimizing

[lgw]

I think the 80/20 rule applies strongly here. Trying to squeeze out the last wasted cycle is, well, a waste these days. However, a little thought goes a *long* way towards performance. Does that algorithm scale properly? Do I really need to allocate that memory dynamically, or is there an easy way to use the stack instead? Is that library call going to do 99% of its work on things unrelated to the results I want? Am I spending resources cacheing something I *know* I'll never use?

Its annoying when people don't bother with even the most obvious performance considerations during design.

Re:Nobody give a fig about optimizing

[groomed]

Bah. n is usually small.

So

[kevinx]

I think the user experience is faster; it's at least richer. You can go ahead and throw windows 3.1 or dos 5 on your machine and watch it scream. Obviously that's now what people want. As long as people demand innovation and new features, they will continue to require more and more power.

You can't eat your cake and have it too.

Re:So

[Flashpot]

But people AREN'T demanding it. It's being foisted upon them. Nobody is DEMANDING that windows-xp is to be delivered on all new peecees except Microsoft.

Change isn't always a given.

[litewoheat]

There's been so mauch change in the past few decades that people keep expecting the same amount of change everywhere. Many people know nothing else. UI as developed by PARC then refined by Apple and Micrsoft hasn't really changed much except for evolutionary steps. There's no revolution coming. Cars have been driven the same basic way since the Model T. Its my firm belief that there will not be revolutionary things such as the printing press, radio, tv, and the Internet coming withing the next 100 or so years. Its time to start refining what we've created rather than look to supplant it.

Re:What about....

[civman2]

You miss the point of the poster. New programs take about as much time to do things on today's systems as old programs did on yesteryear's systems. Old programs are blazingly fast on today's machines.

Usere experience unchaged .. nooo way

[iMaple]

But, in fact, the user experience of performance hasn't improved much over the past 15 years

Now that really depends on what you would call 'user experience'.
Compare a file manager 15 years ols (PC Tools had one right .. for DOS) to the KDE/ Gnome file managers (Ok MC looks the same still :) ).
Compare pine to Thunderbird.(though I still use pine on my old laptop :) )
Compare Usenet clients or say Lynx to Firefox,
Compare Doom 3 to Pac Man .
Comapre the fancy graphics on OS X to Win 3.1 or whatever OS Mac had then

No Sirrr I say the user experience of performance HAS changed. Maybe not directly proportional to the Proceessor speed increase (due to code bloat ?) but still its much much better. Thats my $0.02 .

Re:Usere experience unchaged .. nooo way

[Anonymous Coward]

This is absolutely the point. The user experience hasn't changed in 15 years?

Well 15 years ago was 1990. I was in high school, and my Dad was running a 386sx 16mhz - I think it had maybe 2 meg of ram, and a hdd that was upgraded to 60 meg - of which my dad let me put games on 20. Windows 3 was what we were using, I think. Word processors were just getting the hang of GUI. Wing Commander was the most amazing thing ever. Playing games on your computer involved a whole lot of monkeying with drivers and autoexec.bat / config.sys to get as much free base ram as you could. User experience changed since then? Hell yes.

Charitably, 10 years ago was 1995, and win95 had just come out. Holy crap - multiprocessing! Seebach ends his article suggesting that we open a bunch of programs and see how sluggish our computers get. This "open a bunch of programs" thing only just now - in 1995 - becomes possible in a consumer environment. And, by the way, when was the last time anyone actually ran win95 (i'll give ya osr2!) on a ... say... P133, with oh, 16 or 32 meg of ram (not that this is a '95 era machine...)? You'll wait around for 5 minutes to see that hourglass go away! Run a bunch of applications, then listen to that disk thrash! Seebach's "cranky user" moniker is entirely appropriate. In true curmudgeon style he bitches about today, and compares it to some imagined utopic past. Today I can have windows open all over my 1600 x 1200 desktop. Seriously, like 15 windows is no problem. Not so in 1995.

I'd agree that since maybe the PII - 400 mhz the user experience hasn't lept forward. Why? Because things at that point were happening about as fast as you could want. A PII 400 running a similar era OS - win95 osr2 with 64 - 128 meg of ram on it could bring up the average app as fast as you could want. Since that point I've felt like CPU has had more to offer than wisely written software (of similar era). Even clippy - while stupid - isn't particularly slow. Note that this is on a relatively well maintained machine. All bets are off when Joe User runs "Free Porn Dialer!!!" or turns off his virtual memory or whatever dumbass thing.

Keep in mind that "slow" and "fast" are relative terms. Poor Pete S. double clicks on his Word icon and has time to roll his eyes before the app is ready to do his bidding. In 95 users lucky enough to have a P60 or *gasp* a P90 would order Word - word what, like 2.0? - to start up and would be impressed when it was ready to go in 20 seconds. The fact that people expect things to snap on their computers instantly is a sign of how fast the experience IS today. Back in the late 80's if you knew the name of the file you wanted you could type DIR and almost scan the text fast enough to spot your file as it scrolled the screen. SCROLLING THE SCREEN took a measurable amount of time if your computer had your average trident VGA ISA-bus video board.

All of this, of course, and we don't even need to talk about the difference between EGA porn over the 9600 baud modem from 1990 and the stuff we've got out there now...

Bah. At least, that's the way that I remember it. Maybe I'm the cranky one.

Only Windows

[Apreche]

Have you ever used a completely unpatched windows xp? its fast and stable as hell. It's just full of a zillion security holes. Worse than swiss cheese. If you take it and add all the patches all the way up to the very latest hotfixes, you will notice its slow as hell. That's sans-spyware and additional software. The same thing goes for Office.

Take OSX. Why do people need the dual G5s? Well, its for that fancy gui and shite. If you take that away those boxes are fast as hell. They eat up their cycles with pretty pictures.

And look at Linux, the favorite of slashdot. If you run a "crummy" distro like fedora or lindows or something then your user experience is pretty average. If you use something l33ter like gentoo or debian your user experience will be phenomenal. But whatever performance you've gained by setting up a "real" system you lost all that time when you set the thing up, in the case of gentoo. Or in the case of debian you lost that time learning how to set the thing up.

I tried the new ubuntu linux and this is a very nice distro. It's just as good as a perfect clean debian system, but you don't have to waste time setting it up. The only problem is that all packages that I saw are built for 386. So you're not going to get the most out of your cpu.

If you want your cycles back then run linux without X or run netbsd or something. No matter what you choose, you're going to lose time somewhere. My suggestion is to spend the time to build gentoo, then image a bunch of machines. You get your time back and your cycles, as long as you have multiple machines of similar hardware.

Heh

[FiReaNGeL]

I think that the article writer just realized what a lot of computer buyers don't : CPU speed != more performance, ESPECIALLY when you look at graphical display and Word processing (at least he didn't include "web surfing speed").

Where are my CPU cycles and memory going on my AMD 3500+ and 1Gig 400MHz DDR Ram? Most of the time, nowhere. 1% CPU usage, commit charge 150 megs / 1 gig. Honestly, if you don't use CPU intensive apps, there's a limit to the 'improvement' you can expect in 'graphical display' and 'word processing' speed. But sales rep will tell you otherwise, for sure.

Re:Heh

[hackstraw]

Where are my CPU cycles and memory going on my AMD 3500+ and 1Gig 400MHz DDR Ram? Most of the time, nowhere.

Thats true and untrue, depending on how you look at it. I have a CPU meter on my personal machine and I expect it to be hovering around 0. Right now its just got a little of activity going as I type this because it is spell checking as I type each word. Much less than 10% of the CPU. (I wish OSes came with these things so that people were aware of what was going on with their machine like if its infected with a virus or something, but thats another topic).

I expect it to stay hovering at the 0 mark unless I am doing something. Now, when its not true and when I want good CPU performance is when I launch an app, or do something that requires a little more CPU power like using scp, where it is encrypting or decrypting all of the information that I can put down my network connection as fast as possible. (Or if I forget to leave flash enabled and some cute ad or something is eating 100% of my CPU).

Also its worth mentioning that most people don't need that much CPU power. Look at PowerBook and Mac Mini users. Those are pretty old processors. In the Intel land, anything over 1.5 or so GHz should be more than fine for most people.

Also, most people don't understand bottlenecks. They don't know that CPU processing is about an order of magnitude greater than memory access, and that memory access is about an order of magnitude slower than disk access which in turn is about an order of magnitude slower than network access. I remember when someone 1st got their brand new 486 and I was commenting on how quick it was, and then we were waiting for a file to be copied to or from a floppy and he said "That is fast?".

So the cycles really only get used in bursts for doing something, and it adds to the user experience. With the advent of mulitcore processors, and hopefully power saving by lowering clock speeds when unneeded. I would expect this "bursty" performance to be even more exaggerated. It ads to the user experience. Its only people that really know what they are doing to need sustained CPU processor speed, and for them, nothing is ever fast enough.

Bad software...

[Fleeced]

As a programmer, increased CPU performance means that I churn out lots of quick and dirty code.

Sure, I could spend time making my program more efficient... but why waste a week of my time doing that when the cost of a faster computer is so much cheaper?

Stuff running in the background

[grahamsz]

Generally (at least on my less used windows box) i have most of the following running most of the time....

1) VoIP Client
2) Messaging Client
3) Word Processor
4) Multiple Web Browsers
5) Email Client
6) Probably some graphics or photo editing tool
7) Something playing music

In addition there are various other background processes like desktop indexing, things watching for my digital camera being plugged in, smart start stuff...

Linux is probably worse since i keep Apache and often Tomcat running all the time.

Back in the day, this was never how it was done. You'd optimize config.sys to get the absolute max amount of free conventional memory.

Multitasking has improved to the point that many users probably run close to 100 processes at any point in time..

prstat here says i'm on a system with
Total: 3741 processes, 6739 lwps

Fair enough it's a shared box, but that scale was impossible a decade ago.

Mac OS X

[weave]

An interesting thing I've noticed since I got a laptop running OS X is that it seems every new release and patch for it "improves performance" in some area.

So Apple is bucking the trend, or their first versions of OS X were an inefficient piece of crap and they are just now optomizing it.

Re:Mac OS X

[bonch]

To be honest, part of it was the latter. I don't consider OS X truly usable up until the 10.2 Jaguar release.

It was a totally new operating system replacement for Apple after over a decade of trying, and they rushed it out as quickly as they could just to stay alive (iPods weren't out yet, so they were living off the iMacs). All is forgiven now, especially considering the fantastic technology they had to inherit and integrate. Jaguar was the sweet spot where they finally got it right.

Re:What about....

[Anonymous Coward]

well, then tell me, why does it still take my acrobat reader 10 seconds to launch? 10 seconds for excel? heck, why does it take 30 seconds to launch the OS?

If it launched any faster, you wouldn't see the splash screen and the pretty advertisements + legal boilerplate.

subjective performance...

[bani]

just like users will manage to fill most of the storage space available (no matter how large that may be), user tasks will manage to fill most of the cpu available (no matter how fast the cpu is).

the subjective performance of overall data processing hasn't changed much, but that's just because task complexity has increased as cpu speed increased.

15 years ago, most applications were far less computationally complex than they are today. it has little to do with code bloat.

99 % Waiting

[Jalopy]

As any computer science student knows, most processors spend 99 percent of their cycles waiting. However, now they're waiting much faster.

Re:My memory Usage

[HeghmoH]

Do you have any evidence whatsoever for this? It's not a good idea, and iTunes uses a constant amount of CPU while playing music, indicating a normal buffered, streaming decompression scheme.

Re:My memory Usage

[hasdikarlsam]

So that's why it failed to play a 200MB mp3?

Not very good design, is it?

Thead priorities

[PxM]

Even on systems with limited resources, the lack of proper thread priorities makes the system seem slower than it should be. When MS Word does it's automatic spell check, there shouldn't be a noticeable drop in UI performance (lag between hitting a key and a seeing letter appear) if the CPU intensive tasks ran in the background. The same applies to how FF lags badly when I open a PDF inline or how Explorer's interface locks and doesn't refresh properly when it's previewing a video file or reading from a slow CD drive. The same should apply to background tasks like virus scanners, but they tend to have the added overhead of using IO which is harder to multitask smoothly.

If the program was designed so that the functional part of the application had the highest priority and the eye candy and fluff ran in the background, the user wouldn't be as annoyed with the high peak performance requirements of certain programs.
--
Free iPod? Try a free Mac Mini
Or a free Nintendo DS
Wired article as proof

My favorite theory...

[merlin_jim]

Is that it is (like nearly every other system out there) an economical system; it's all about supply and demand.

I this case there's a supply of plenty of clocks. There's an (existing) demand for a certain level of performance; if the supply outstrips that demand, then the supply is devalued, and consequently the programmers don't spend as much time conserving that resource.

Or to put it another way, programs behave like a gas with respect to responsiveness and user expectation; they expand to fill the available space.

Or to reword it another way (quoting from the article): computers are, in fact, doing more than they used to. A lot of the things computers do are fairly subtle, happening beneath the radar of a user's perception. Many functions are automatic and, as discussed in last month's column, you could probably do without some of them.

Re:My memory Usage

[Anonymous Coward]

From a technical point of view, yes, a very typical Apple design. Good enough to fool the average user but bad enough to really, really fucking annoy the user with atypical, demanding, or stringent expectations and needs.

Its not bloat if you derive utility from it

[Ars-Fartsica]

I am tired of seeing this rolled out naively again and again. I like smooth fonts, multimedia support, device management, hires icons, a little bit of eye candy etc etc.

I derive utility from them and my hardware can handle it. This is not bloat.

Re:Its not bloat if you derive utility from it

[EmbeddedJanitor]

What is bloat though is the shitty code behind these. Nice bloaty STl C++ class implmentation that does, in 200 line, 5kbytes and 5000 cycles what a tight bit oc C can do in 20 lines, 100 bytes and 200 cycles.

Bloating isn't just in RAM, CPU and diskspace. It's now happening to other resources like network bandwidth too. Ten to 15 years ago I could do a lot of very useful stuff with a 1200baud dial up. Now 50k dialup is pure crap for many purposes, mainly because of bloat.

The Future of computing

[neomage86]

5 years from now some *nix (most likely the latest iteration of MacOS) on parallel cell CPUs. 10 years, a unified virtual machine so applications will be OS independent(I think by Microsoft) 15 years Computing will be a commodity. I will pay $N a month for computer usage (based on what specs I want my computer to have). There will be ubiquitous dumb terminals (everywhere from home, to work, to school) each will have a massive (by today's standards) data pipe. You authenticate and it will become your computer. Your desktop of choice, your files, your preferences. There will be few massive datacenters, so everything is amazingly scalable and centralized. If I want to do video editing, I will have access to supercomputers worth of power while I need it, and while word processing I will use almost nothing. Right now, it looks like Google will run these datacenters.

Vested Interests

[benjamin_pont]

Don't you think that if we all had zippy computers with slim, efficient operating systems and applications that made modest use of resources, and had only the features people wanted, then there would be a lot of bankrupt technology companies and unemployed programmers since no one would be upgrading their systems (much)?

* Excerpt from Aldous Huxley's Brave New World *

"We condition the masses to hate the countryside," concluded the Director. "But simultaneously we condition them to love all country sports. At the same time, we see to it that all country sports shall entail the use of elaborate apparatus. So that they consume manufactured articles as well as transport. Hence those electric shocks."

"I see," said the student, and was silent, lost in admiration.

By the way, current number of mouse-clicks to configure viewing an MS Outlook sender in a given color:

17

Don't $top that fat, gravy-train from rolling! Keep the bloatware coming!

Get a clue!

[www.sorehands.com]

Only a small portion of code bloat is actually useful features. But it is more related to marketing/review checkbox features -- features that are not really used, but it is added only because it can be shown as being there for marketing.

The other part is that as time goes on many programmers are lazy and they don't think though the code and optimize the code because it is fast enough. There are not as many real programmers that take pride in their code and make it efficient.

It's not just that...

[bonch]

A lot of people like to wax nostalgic about the good ol' days when things were faster, all the while ignoring the featuresets implemented today. The fact is, yesteryear's code wasn't as advanced or featured. A word processor from ten years ago may have been pretty damned fast, but today's word processors load dictionaries to spellcheck your document and grammar as you type and import graphics.

Windows 3.1 may have been incredibly snappy, but it also lacked propery memory protection, wasn't 32-bit addressed, and didn't provide an intuitive interface. Also, more advanced typography (anti-aliasing) and filesystem indexing services like Spotlight come into play, as well as all the important system daemons running in the background that are now considered stock.

It's not that things aren't getting any faster because of bloat. It's because as power increases, the ability to add new modern features to the original experience is utilizing that extra power, just as it should.

Re:It's not just that...

[The Bungi]

Windows 3.1 may have been incredibly snappy

I see your point, but let's put things in perspective - Windows 3.x was not snappy on common hardware available in that day, which was pretty much a 486DX2-50/66 with 8MB (average) of RAM and slowish IDE HDs. It really wasn't until the Pentium 60/66 became more mainstream (and I managed to get my hands on one) that 3.x actually seemed damn fast. This was especially true for WFW, which was rather slow on a DX33, but OTOH it actually got better if you threw 32MB or RAM at it, whereas the normal version of Windows stopped noticing once you went past 16-24MB.

Don't get me wrong, I'm not saying it was unusable. It just wasn't particularly snappy.

It's always been this way - when Windows XP came out the majority of people were still running PIII's in the 450-900MHz range; it wasn't until you got a P4 that things started looking rosy. XP/2003 on a 3GHz P4 with HT are positively snappy. But as long as Intel continues to deliver...

Re:Fun fact!

[Johnno74]

Nope, wrong. Task manager reports working set, which is whatever memory windows THINKS the process is using at the moment.

Reporting the total memory allocated by an app is meaningless, because much of that memory includes things like memory-mapping DLLs that are only loaded once, and shared amongst all processes using them.

Windows constantly trims pages from the working set of each process that haven't been accessed in a while and pages these out to disk.

When you click the minimize button windows assumes that the app (or at least large parts of it) are going to be inactive for a while, so it tries to remove as many pages from the working set as possible - hugely reducing the "memory usage" reported in task manager. Sometimes it is a little too agressive in trimming pages from the working set, and the "memory usage" immediately climbs a little again, as the app accesses memory pages that windows thought it could page out.

Try it with an app like MS Word that has background threads that constantly check spelling and stuff, and you'll see the working set goes up for a few seconds after you minimize it.

Re:Fun fact!

[Coryoth]

I'm sooo sick of people looking in Task Manager then saying how much an application sucks because of how much "memory" it uses. For the most part, memory is not a factor.

You think that's annoying? You should try having to put up with all people who complain "X is sooooo bloated" because they looked at memory usage on top but have no idea how it is calculated for X, nor what the figure really means. Every single article that has any mention of X11 gets at least 15 posts all saying X is bloated based on top, and every time there are a bunch of people who give long and careful explanations as to why the figures in top are the way they are. And then the next article comes along and a whole bunch of people decide to demonstrate exactly how little they've learned...

Jedidiah.

Microsoft Called....

[raehl]

They said they were never your cycles in the first place.

Let's go back to assembly, non-PNP, and text mode!

[WoTG]

I hate talk about bloat, slowness, and lazy programmers. For the same tasks, computers are far easier to use today than even 5 years ago. Never mind the fact that we can also do about 10 times as many different tasks now, than we did when I first started using PCs 10-15 years ago.

Yes, if I could find a floppy drive, and get a dos boot disk to boot, I could theoretically run a wickedly fast instance of WordPerfect 5.1. I wouldn't be able to surf the web, send email, listen to MP3's, work wirelessly, or work with graphics though -- and yes, graphics (e.g. diagrams) do have a proper place in day to day work!

Do people even remember the non-PNP days? IRQ's, IO Ports and the rest? Non-multitasking? Non-Memory protected (i.e. complete OS crashes from app errors?). These issues didn't seeem so bad back then since "that's the way it was", but now, I dread ever having to deal with those limitations again. Futzing with IRQ's for an hour just to make a modem stop locking up a PC is not my idea of productivity.

Hardware is cheap. Time isn't. I just hope we keep finding more ways to make my use of computers even easier.

ummm... kernel compile?

[rsw]

I don't know about you guys, but back when I was running linux on a 33 MHz 80386, my kernel compiles went overnight. Now they take, oh, ten minutes.

I'd say that's an improvement, wouldn't you?

What kind of optimization?

[DeadVulcan]

nobody cares about optimizing code anymore.

You can optimize in many different ways: for run-time performance, maintainability, extendibility, usability, compatibility, and probably a bunch of other ways I can't think of just now.

Many of these are at odds with each other. And since computers are getting faster, I think it's perfectly reasonable to start trading off run-time performance with some of these other things.

Re:My memory Usage

[TheGavster]

Two things. First, could it possibly be under Windows? Try minimising it and tell us again.
Well, that's just shite interface programming. Other apps don't suck 60MB to display a window.

Next. To put your question differently "Why does Matlab uses 300Mb just to add two numbers?" Because it is intended for more than that?
Matlab is designed to perform complicated calculations and analyses, so simple integer addition is an inappropriate application. iTunes is designed to play media files, so playing an MP3 is an appropriate application. Your point is void.

TFA is a piece of junk

[patio11]

Something tells me the author has no clue how virus scanners actually work. Unless those games loading image files are loading them in a known macro-virus-prone format (and, I can say without fear of contradiction, WoW isn't parsing textures from .docs), the gigantic resource files will *never be passively scanned* because they are *incapable of being executed*.

You can verify this for yourself. Get a file containing a virus signature (find an old trojan somewhere). Disable your virus scanner, copy it to an arbitrary place on your machine, rename it to .dat, and turn on the virus scanner. Your virus scanner won't pick up the file unless you've set it to active scan everything on certain intervals. Leave it there for a few weeks and it will never be picked up by the passive process because the passive process *doesn't worry about non-executable data*.

CPU usage caused as a result of automatic spellchecking is another howler. Its one thread, which will spend most of its time blocking for IO (thats you, chief!) Typical usage patterns will see no slowdown due to the checking thread -- open up your favorite resources manager and start banging away at MS word, unless you type faster than God or start dumping copy-pasting or macro magic into the file your CPU usage won't even budge. The algorithms, by the way, give maximum priority to the user input handling threads and close to minimum for the background checks, because word processor developers know the perception of speedy response is one of the key features of their product. As a result, they'll generally not start spell-checking a copied block until the CPU is otherwise underutilized (i.e. blocking on user IO).

I don't agree with this analysis...

[Doppler00]

Okay a few comments here:

On the downside, antialiasing sucks up a lot of processing power.

I don't see how this is possible. After scaling, the font should be properly cached as a bitmap. Thus, it would not use very much CPU after being rendered into memory. Has the author of this article actually benchmarked this to show how "slow" antialised fonts are? Or is he making stuff up.

Visual effects like drop shadows behind windows and menus, transparent menus and effects, and real-time effects also consume a lot of processing power. The catch for computer makers is that most users expect them.

Transparency and shadows should be accelerated by the video card. This should result in minimal CPU cycles being used for this. But really, if this eye candy is bothering you, just disable these features and you're computer is 100x faster again. Often, it's not the rendering of these effects that makes the computer slow, but the fact that an animation HAS to occur before something else happens.

support code that no one really understands
What does he mean by this? He's equating more manageable code with slower code. Is he implying that complex, unmaintainable code is faster? I've never seen that to be the case.

Each of these programs pre-loads its own shared libraries

This is something I'll agree with. Each time I install a new driver for some device on my computer it requires a service on startup to use about 5MB-10MB of memory. My guess is that the driver developers don't think it's worth their time to optimize memory usage.

The reason why some applications are slower is simply because it was never in the requirements for them to be faster. It's just a quality control issue. There is a trade off between time to market and software performance. The performance usually is the last thing considered in a product (just get it working first).

Re:My memory Usage

[mcrbids]

Only on Slashdot: a senseless reply to sarcastic humor gets modded "insightful"...

Contrary to a recent troll...

[SanityInAnarchy]

Linux actually is incredibly fast, especially if you tweak it. The only time my Linux ever feels slow is during boot, and that's my desktop -- my laptop "boots" (from Software Suspend) in less than 40 seconds, and it's 1 ghz, 256 megs of ram, 15 gig hard drive that can't be more than 5400 rpm...

You're right, new features get added, but I got decent performance on a 200 mhz workstation with Gentoo. It's only when we got into the things that cannot really be optimized (compiling, 3d gaming) that I had to by my current box.

The only place Linux ever feels slow is the same place Windows does -- starting programs, loading stuff from disk -- only less so, with less ram. And it's still much faster, because I can tweak Linux much more easily than Windows.

and all that spyware

[Anonymous Coward]

Dont forget the 30 different peices of spy ware all trying to advertise for you and doing 50000 registry mods per second to try to stay alive.

Oh btw, install Win2k and see how damn fast it is compared to XP.